The modifications of Mercury surface induced by solar wind ions

Università degli Studi di Catania

Scuola Superiore di Catania

International PhD in

Nuclear and Particle Astrophysics

XXVI cycle

The modifications of Mercury surface

induced by solar wind ions

Ivana Sangiorgio

Coordinator of PhD

Prof. Stefano Romano

Tutor

Prof. Francesca Zuccarello

Co-Tutor

Di terra i profumi, lui, tutti inspirò, posate e canzuni, in gran chiasso, accordò. Dei pesci i costumi, in silenzio, spiò, con ganci e buttùni ogni legge sfidò. La “Banca” in volumi, il greco e i Maccarrùni,

‘u iàttu e i “pirsùni”,

mugghieri e vasùni… … Giostra allegra di fiabe lì, per te, s’inventò e cullarti coi lumi e L’Amore ora può.

Alla mia famiglia

I

i

I

NDEX

Introduction 1

Chapter 1

What we know about the planet Mercury 5

1.1 Mercury and space weathering 5

1.2 Missions addressed to Mercury 8

1.2.1 Mariner 10 8

1.2.2 MESSENGER 10

1.2.3 BepiColombo 15

1.3 Terrestrial planets: Mercury 19

1.4 Magmatic rocks and minerals 21

1.5 Mercury and the Moon 24

1.5.1 Lunar soil modifications 24

1.5.2 Similarity with the Moon? 27

1.6 Experiments and modeling in literature 30

1.6.1 Reflectance spectra and compositional heterogeneity 30

1.6.2 Exosphere-surface relationship 32

1.7 Outline: which are, then, Mercury’s surface building blocks 36

Chapter 2

Experimental techniques 39

2.1 Experimental analytical techniques:

Laboratory for Experimental Astrophysics (LASp) 39

I

ii

2.2.2 The spectrophotometers 43

Section A: IR spectrophotometer and data 43

Section B: UV-Vis NIR spectrophotometer and data 48

2.2.3 The ion implanter 53

2.3 Materials and experimental procedure 56

2.3.1 Etnean samples 56

2.3.2 Other silicate materials 61

Chapter 3

Laboratory spectra of Mercury analog materials 63

3.1 UV-Vis-NIR reflectance spectra of Etnean basalts 63

3.2 IR reflectance spectra of Etnean basalts 70

3.3 Irradiated etnean samples 74

3.4 Other silicate spectra 79

3.4.1 Silicate films on silicon wafers 79

3.4.2 Silicate rocks spectra 82

Chapter 4

Observations and comparison with laboratory spectra 85

4.1 Mercury spectra 85

4.2 Sulfides 94

4.3 Comparison between observational and laboratory spectra 97

4.4 Discussion 105

Bibliography 109

I

1

I

NTRODUCTION

The space through which Solar System bodies move is not empty. The Interplanetary Medium fills it with dust, photons, meteorites of different size, cosmic rays and a mixture of electrons, protons, heavier ions and magnetic fields, the solar wind, streaming radially outward in all the directions from the Sun at supersonic speeds of hundreds of kilometers per second.

How the Interplanetary Medium interacts with planets depends on whether they have magnetic fields or not.

Airless bodies, such as Mercury, are continuously exposed to the above mentioned bombardments that can induce on their surfaces sputtering, implantation, photo or thermal desorption, vaporization, erosion. These processes which alter physical and optical properties of the surfaces and contribute to the mass loss of the bodies, constitute the so called space weathering.

Mercury is the innermost Solar System planet, with the highest orbital eccentricity (e=0.205) and complete three rotations around its axis for every two revolutions around its orbit. These peculiarities give to Mercury a great thermal excursion (~700 K) between Sun-exposed and the “dark” sides and make this planet a very interesting subject for the astrophysical community because information regarding its principal constituents is not complete.

Despite its position in the Solar System, some ground observations of Mercury, in the visible, near and mid infrared spectral region (0.5-25 µm), have been made and have revealed, with a few uncertainties, a heterogeneous mineralogical surface composition and the presence of sodium, potassium and calcium.

I

2

The greatest contribution to the knowledge of the planet has been obtained thanks to space missions because they have acquired, and are still acquiring, an huge quantity of resolved images and data. The future joint European-Japanese space mission, BepiColombo, will add complementary information and new laboratory experiments are also requested for testing its instruments, prepare and validate the results.

The work in this thesis arises from a scientific collaboration between Osservatorio Astrofisico di Catania of INAF (Istituto Nazionale di Astrofisica), and IAPS-INAF (Istituto di Astrofisica e Planetologia Spaziali) in Rome, in which operate the Principal Investigators of two experiments (ISA, Italian Spring Accelerometer and SERENA, Search for Exospheric Refilling and Emitted Natural Abundances) of the BepiColombo mission. The aim of the present work is to characterize some Mercury analog materials in preparation to ELENA (Emitted Low-Energy Neutral Atoms) experiments aboard BepiColombo.

The candidate materials, here chosen on the basis of existing data in literature, are old and fresh erupted etnean basalts, together with other silicate rocks (sodalite, nepheline, jadeite) containing sodium and potassium.

The variations in the spectral reflectance of these candidate Mercury materials, before and after irradiation with energetic ions (200 keV H+ _{and Ar}+_{) aimed at simulate cosmic ion}

bombardment, have been studied, with the intent to be compared with observational Mercury spectra.

Some possible analogies have been tested operating an opportune scaling of the laboratory ion energy and fluences, in order to estimate the correspondent exposure time scale at Mercury.

Our Mercury analog materials seem to be the best candidates so far found, even if a wider study on different materials is needed.

Furthermore the changes in the spectral reflectance of irradiated laboratory samples confirm that space weathering on Mercury surface produces relevant spectral variations of the surface reflectance.

This work starts from Chapter 1 with a description of the state of the art of Mercury (obtained thanks to space missions Mariner 10 and MESSENGER and thanks to some experiments and modeling), a presentation of the future mission BepiColombo, and a definition of terrestrial planets and magmatic rocks.

I

3

In Chapter 2 the detailed experimental instrumentations, the working procedures concerning the sample preparation and their characteristics are presented, while the UV-Vis-NIR and IR reflectance spectra before and after irradiations of the Mercury analog laboratory samples are reported in Chapter 3.

Chapter 4, the conclusive, is dedicated to the presentation of Mercury spectra so far obtained from ground and MESSENGER satellite in the 400-1000 nm spectral region. In this chapter we present and discuss the comparison between observational and reflectance spectra of our laboratory sample.

C

1

W

M

5

C

HAPTER

1

W

HAT WE KNOW ABOUT THE PLANET

M

ERCURY

Mercury was named after the Roman god Mercury (who was regarded to be the speedy messenger between the gods) because the planet seemed to move so quickly in the sky. It slips away to a full comprehension yet, due to the extreme conditions to which Mercury’s environment is exposed and some of its peculiarities and anomalies, however, make this planet a subject of special scientific interest.

In this chapter we will try to trace the main features on the basis of existing data in the literature.

1.1 Mercury and space weathering

The surface of the airless bodies in the Solar System is endlessly struck by several energetic processes. Cosmic ions, solar wind and coronal mass ejections, photons, micrometeorite impacts can induce sputtering, implantation, photo or thermal desorption, vaporization,

C

1

W

M

6

cause the erosion of the surfaces. These processes, which alter physical and optical properties of the surfaces and contribute to the mass loss of the bodies, constitute the so called space weathering (Baratta et al., 2011).

Spacecraft observations permit to assert that Mercury’s crust was shaped by volcanism, deformation and impact cratering and the presence of albedo and color variations is related to compositional heterogeneities (Denevi et al., 2009).

The most important findings are summarized in Table 1.1 and will be described below. Mercury’s middle orbit is 0.387 AU (1 AU, Astronomical Unit = average Earth-Sun distance=1.496×1011 _{m), with an orbital period of about 88 terrestrial days, and a rotation}

period of 59 terrestrial days. These orbital characteristics, together with a very high eccentricity (e=0.205), gives to Mercury an extremely variable temperature, ranging from 770 K at the perihelion on the Sun exposed surface, to 90 K at the aphelion, in the “dark” side.

Table 1.1- Mercury’s parameters (Milillo et al., 2005).

Mercury ☿

Radius 2240 km (0.38 REarth)

Mass 3.3022×1023 _{kg (0.055 Earth mass)}

Aphelion 69816900 km (0.466 697 AU)

Perihelion 46001200 km (0.307 499 AU)

Gravity acceleration 0.377 g

Eccentricity 0.205

Mercurian day ~ 59 Earth days

Mercurian year ~ 88 Earth days

Mean density ~ 5.43 g∙cm-3 _{(~ 0.98 Earth density)}

Uncompressed density _{~ 5.3 g∙cm}-3

Temperature range 770 K (perihelion day side) 90 K (aphelion night side) Because of its vicinity to the Sun, its relatively small dimensions and its gravity, Mercury is not able to maintain a proper atmosphere, but only a tenuous gaseous envelope in which molecules cannot collide (the exosphere).

C

1

W

M

7

This region, together with the surface, is continuously modified by many processes, as described below.

Thanks to Mariner 10, the first space mission to Mercury, we knew about the existence of an internal magnetic field, not yet well established, but able to reject a great part of the solar wind ions. Solar wind ions are composed of 95% protons, about 5% of α-particles and traces of heavier nuclei. Nevertheless, in the regions of the magnetic reconnection between Mercury magnetic field lines and those of the interplanetary magnetic field (cusps), ions can penetrate, impact the surface and can create neutral and also ionized particle emission (sputtering). A part of the ions penetrated is diffused and circulate around the closed line of the magnetic field, or falls down in the surface, according to the mass, in the middle latitudes; another part of these ions exchange the charge with the thermal atoms emitted in the exosphere, producing Energetic Neutral Atoms (ENA).

In addition to sputtering, particles can be removed from the surface by photon stimulated desorption (PSD), the micrometeorite impact vaporization (MIV) and the thermal desorption (TD) of light atoms (H, He).

The kinetic energy of the released particles depends on the process involved and is below a few eVs for photon-stimulated desorption (PSD) and thermal evaporation, a few eVs for micrometeoroid impacts, and can reach higher values (few tens of eV) for ion sputtering (Milillo et al., 2005).

The Sputtered High Energy Atoms (SHEA), the Energetic Neutral Atoms (ENA), together with the accelerated ions can escape the planet attraction and contribute to the mass loss of the surface and to the erosion of the exosphere.

Mercury surface, its exosphere and magnetosphere, under the influence of the solar wind and the interplanetary field, are then a strongly coupled dynamic system mostly ruled by the interactions between the neutral/ionized gas particles with the surface and the magnetospheric plasma.

As previously mentioned, the major contribution to the knowledge of the planet Mercury has been given by space missions. In the next section these will be presented in detail.

C

1

W

M

8

1.2 Missions addressed to Mercury

Because of its position in the Solar System, it has been difficult to observe Mercury from Earth.

Space missions have rebuilt the knowledge of the planet Mercury because they have allowed to acquire resolved images and data in a wide range of the electromagnetic spectrum.

The missions addressed to Mercury are the past Mariner 10, the still actual MESSENGER, and the future BepiColombo.

1.2.1 Mariner 10

Mariner 10 was an American robotic space probe launched by NASA on November 3, 1973, to fly by the planets Mercury and Venus.

It was the first spacecraft to visit Mercury and to use the gravitational pull of one planet (Venus) to reach another (Mercury), and the first spacecraft mission to visit two planets. Was Michael Minovitch, a university student who worked in the summer at the Jet Propulsion Laboratory of NASA, who in 1963 developed the concept of gravitational propulsion. He thought of using the slingshot effect with propulsive purposes in a systematic way, resulting in a net reduction of propellant required for missions addressed to several planets.

In 1970, Giuseppe Colombo (known also as Bepi Colombo), Professor of Applied Mechanics at the Faculty of Engineering of Padua, proposed an important modification to the Mariner 10 path. Colombo found that the orbital period of the probe after the fly-by of Mercury would have coincided with twice the period of revolution of the planet itself and suggested to use such resonance to program multiple flybys of Mercury.

Thanks to these important contributes, the spacecraft was able to fly by Mercury three times in a retrograde heliocentric orbit, using the solar radiation pressure on its solar panels and its high-gain antenna as means of attitude control during flight, and returned the first-ever close-up images of Venus and Mercury and data on the planets.

The probe was launched onboard a carrier Atlas/Centaur, at Cape Canaveral Air Force Station, Florida.

C

1

W

M

9

Mariner 10 instruments (shown in Fig. 1.1, on the left panel) included: twin telescope/cameras, with digital tape recorder, which took images alternately, every 42 seconds, and sent them directly to Earth; an ultraviolet occultation spectrometer, designed for the detection of a possible atmosphere of Mercury which was operating only during the flybys of the planet; an ultraviolet airglow spectrometer, for the detection of the radiation emitted by the atmosphere; an infrared radiometer, which measured the extent of the thermal radiation emitted by the upper layers of the clouds of Venus and from the surface of Mercury at the wavelengths of 11 and 45 micrometers; two magnetometers, in order to distinguish the magnetic disturbance generated by the sensor itself from the magnetic field of the interplanetary plasma; two charged particle telescopes, sensitive to charged particles (protons, α-particles, electrons) and operating at different energy intervals; a scanning electrostatic analyzer and a scanning electron spectrometer to analyze the distribution of the solar wind in interplanetary space and assess their interaction with Venus and Mercury. The probe carried also a motor driven high-gain dish antenna and a low-gain omnidirectional antenna.

Instruments on-board the spacecraft measured the atmospheric, surface, and physical characteristics of Mercury and Venus. Experiments included television photography, magnetic field, plasma, infrared radiometry, ultraviolet spectroscopy, and radio science detectors. However, some technical problems allowed to operate only a part of these experiments.

An experimental X-band, high-frequency transmitter was flown for the first time on this spacecraft.

The spacecraft passed Venus on February 5, 1974 and crossed the orbit of Mercury on 29 March 1974, at a distance of about 704 km from the surface. A second encounter with Mercury, when more photographs were taken, occurred on 21 September 1974, at an altitude of 48069 km. A third and last Mercury encounter at an altitude of 327 km, with additional photography of about 300 frames and magnetic field measurements occurred on 16 March 1975. Engineering tests were continued until 24 March 1975, when the supply of attitude-control gas was depleted and the mission was terminated.

Owing to the geometry of its orbit – its orbital period was almost exactly twice Mercury's – the same side of Mercury was sunlit each time, so it was only able to map 40–45% of

C

1

W

M

10

Mercury’s surface, taking over 2800 photos (a mosaic of a part is represented in Fig. 1.1, on the right panel).

It was confirmed that Mercury has a tenuous atmosphere consisting primarily of helium and a cratered, dormant Moon-like surface was shown in the images. Mercury was shown to have a small magnetic field and a relatively large iron-rich core.

Fig. 1.1 – Left panel: an image of the Mariner 10 spacecraft with its instrumentations. Right panel: a mosaic photo taken by the Mariner 10 spacecraft during its first approach to Mercury. The mosaic consists of 18 images taken at 42 s intervals during a 13 minute period when the spacecraft was 200000 km from the planet.

1.2.2 MESSENGER

MESSENGER (MErcury Surface, Space ENvironment, GEochemistry, and Ranging) is a robotic NASA spacecraft orbiting the planet Mercury, the first spacecraft ever to do so. It was launched aboard a Delta II rocket in August 2004 to study Mercury's chemical composition, geology, and magnetic field. It made a first flyby in January 2008, followed by a second flyby in October 2008, and a third flyby in September 2009.

The spacecraft flew by Earth once, Venus twice, and Mercury itself three times, allowing it to decelerate relative to Mercury using minimal fuel. MESSENGER successfully entered Mercury's orbit on March 18, 2011.

C

1

W

M

11

MESSENGER's dual-mode, liquid chemical propulsion system is integrated into the spacecraft's structure to make economical use of mass. The structure is primarily composed of a graphite epoxy (resins) material, which provides the strength necessary to survive launch while offering lower mass. Two large solar panels, supplemented with a nickel-hydrogen battery, provide MESSENGER's power.

The spacecraft was designed and built at the Johns Hopkins University Applied Physics Laboratory. Science operations, managed by Dr. Sean Solomon as principal investigator, and mission operations are also conducted at JHU/APL.

On March 17, 2012, having collected close to 100000 images, MESSENGER ended its one-year primary mission and entered an extended mission scheduled to last until March 2013. Now MESSENGER is exhausting its fuel and gradually reducing its orbit, until it will impact the planet. A picture showing the MESSENGER satellite current position is shown in Fig. 1.2.

Fig. 1.2– An image showing the surface of Mercury as seen from MESSENGER’s current position as well as the MESSENGER spacecraft’s current orbit and orientation. Positions of stars with magnitude 5 or brighter are shown. The depiction of Mercury’s surface includes images acquired by the Mariner 10 and MESSENGER spacecraft.

C

1

W

M

12

The mission's primary science objectives included: determining the surface composition of Mercury, characterizing the geological history of the planet, determining the precise strength of the magnetic field and its variation with position and altitude, investigating the presence of a liquid outer core by measuring Mercury's libration, determining the nature of the radar reflective materials at Mercury’s poles, investigating the important volatile species and their sources and sinks on and near Mercury.

To reach these objectives, MESSENGER was supplied with several instruments aboard (listed in Table1.2 and indicated in Fig. 1.3, left panel): a Mercury Dual Imaging System (MDIS), a Gamma-Ray and Neutron Spectrometer (GRNS), a X-Ray Spectrometer (XRS), a Magnetometer (MAG), a Mercury Laser Altimeter (MLA), a Mercury Atmospheric and Surface Composition Spectrometer (MASCS), an Energetic Particle and Plasma Spectrometer (EPPS), Radio Science (RS).

Table 1.2- MESSENGER instruments and the acronyms used.

Name Acronym

Mercury Dual Imaging System: -Narrow-Angle Camera

-Wide-Angle Camera

MDIS: -NAC -WAC

Gamma-Ray and Neutron Spectrometer GRNS

X-Ray Spectrometer XRS

Magnetometer MAG

Mercury Laser Altimeter MLA

Mercury Atmospheric and Surface Composition Spectrometer: Ultraviolet and Visible Spectrometer

Visible and Infrared Spectrograph

MASCS UVVS VIRS Energetic Particle and Plasma Spectrometer:

-Energetic Particle Spectrometer -Fast Imaging Plasma Spectrometer

EPPS: -EPS -FIPS

C

1

W

M

13

Mercury Dual Imaging System (MDIS) includes two CCD cameras, a narrow-angle camera (NAC) and a wide-angle camera (WAC) mounted to a pivoting platform. The camera system provides a complete map of the surface of Mercury at a resolution of 250 meters/pixel, and images of regions of geologic interest at 20–50 meters/pixel. Color imaging is possible only with the narrow-band filter wheel attached to the wide-angle camera (in Fig. 1.3, right panel, is shown the global coverage of Mercury surface obtained with this instrumentation).

Gamma-Ray and Neutron Spectrometer (GRNS) is able to detect gamma rays and neutrons that are emitted by radioactive elements on Mercury's surface or by surface elements that have been stimulated by cosmic rays. It is used to map the relative abundances of different elements and help to determine if there is ice at Mercury's poles, which are never exposed to direct sunlight.

Gamma rays and high-energy X-rays from the Sun, striking Mercury's surface, can cause the surface elements to emit low-energy X-rays.

X-Ray Spectrometer (XRS) was thought to detect these emitted X-ray spectral lines and to measure the abundances of various elements in the materials of Mercury's crust, as magnesium, aluminum, sulfur, calcium, titanium, and iron, in the 1-10 keV range. It can map mineral composition within the top millimeter of the surface on Mercury.

Magnetometer (MAG) measures the magnetic field around Mercury in detail to determine the strength and average position of the field, and search for regions of magnetized rocks in the crust.

Mercury Laser Altimeter (MLA) provides detailed information regarding the height of landforms on the surface of Mercury by detecting the light of an infrared laser as the light bounces off the surface.

Mercury Atmospheric and Surface Composition Spectrometer (MASCS), combining an ultraviolet spectrometer and infrared spectrograph, is able to measure the abundance of atmospheric gases around Mercury and to detect minerals in its surface materials.

The Ultraviolet and Visible Spectrometer (UVVS) can determine the composition and structure of Mercury’s exosphere and study its neutral gas emissions. It also searches for and measures ionized atmospheric species. Together these measurements will help researchers understand the processes that generate and maintain the atmosphere, the connection between

C

1

W

M

14

surface and atmospheric composition, the dynamics of volatile materials on and near Mercury, and the nature of the radar-reflective materials near the planet’s poles.

The Visible and Infrared Spectrograph (VIRS) measures the reflected visible and near-infrared light at wavelengths diagnostic of iron and titanium-bearing silicate materials on the surface, such as pyroxene, olivine, and ilmenite.

Energetic Particle and Plasma Spectrometer (EPPS) measures the composition, distribution, and energy of charged particles (electrons and various ions) in Mercury's magnetosphere using an Energetic Particle Spectrometer (EPS). It measures also the charged particles that come from the surface using a Fast Imaging Plasma Spectrometer (FIPS).

Radio Science (RS) use the Doppler effect to measure very slight changes in the spacecraft's velocity as it orbits Mercury. This allows to study Mercury's mass distribution and its gravity, together with variations in the thickness of its crust.

Fig. 1.3– Left panel: a sketch of the MESSENGER satellite and its instrumentations. Right panel: global coverage obtained with MESSENGER. The globe on the left was created from the MDIS monochrome surface morphology base map campaign. The globe on the right was produced from the MDIS color base map campaign. Each map is composed of thousands of images, and the color view was created by using 3 of the 8 color filters acquired (1000, 750, and 430 nm wavelengths are displayed in red, green, and blue, respectively). Date acquired: February 22, 2013- NASA/Johns Hopkins University Applied Physics Laboratory/Carnegie Institution of Washington.

During its stay in Mercury orbit, MESSENGER's instruments have yielded significant data and have achieved many of the goals that had been set, including a characterization of

C

1

W

M

15

Mercury's magnetic field and the discovery of the presence of O-H bonds possibly due to water ice at the planet's north pole.

In the following pages several of these achievements will be treated and described and they will be a constant reference point in this thesis work.

1.2.3 BepiColombo

BepiColombo is named after the mathematician, astronomer and engineer Giuseppe (Bepi) Colombo (Padua, 1920-1984) who discovered the coupling between rotation and revolution of Mercury and carried out a campaign to promote the Italian space research, as previously outlined.

This is the first European mission addressed to Mercury, and arises from the need to understand the mysteries of the planet, not yet completely resolved.

BepiColombo is a joint mission between ESA (European Space Agency) and the Japan Aerospace Exploration Agency (JAXA), executed under ESA leadership, was previously targeted for launch in July 2014, and is now planned for launch in a window opening in August 2015, because of a risk mitigation strategy and budgetary constraints.

The science mission will consist of two separate spacecraft that will orbit the planet (whose orbits are represented in Fig. 1.4, right panel, and compared with that of the MESSENGER satellite). ESA is building one of the main spacecraft, the Mercury Planetary Orbiter (MPO), and the Institute of Space and Astronautical Science (ISAS) at the JAXA will contribute the other, the Mercury Magnetospheric Orbiter (MMO). The MPO will study the surface and internal composition of the planet, and the MMO will study Mercury's magnetosphere. The instruments aboard the two units are reported in Tables 1.3 and 1.4.

C

1

W

M

16

Table 1.3- MPO instruments and the acronyms used.

BepiColombo Laser Altimeter BELA

Italian Spring Accelerometer ISA

Mercury Magnetometer MERMAG

Mercury Thermal Infrared Spectrometer MERTIS

Mercury Gamma ray and Neutron Spectrometer MGNS

Mercury Imaging X-ray Spectrometer MIXS

Mercury Orbiter Radio science Experiment MORE

Probing of Hermean Exosphere by Ultraviolet Spectroscopy PHEBUS

Search for Exosphere Refilling and Emitted Neutral Abundances (Neutral and ionised particle analyser):

- Emitted Low-Energy Neutral Atoms

- STart from a Rotating FIeld mass spectrOmeter - Miniature Ion Precipitation Analyser

- Planetary Ion Camera

SERENA: -ELENA -STROFIO -MIPA -PICAM Spectrometers and Imagers for MPO BepiColombo Integrated Observatory

System (High resolution and stereo cameras, Visual and NIR spectrometer)

SIMBIO-SYS

Solar Intensity X-ray Spectrometer SIXS

Table 1.4- MMO instruments and the acronyms used.

Mercury Magnetometer MERMAG-M/MGF

Mercury Plasma Particle Experiment MPPE

Plasma Wave Instrument PWI

Mercury Sodium Atmospheric Spectral Imager MSASI

Mercury Dust Monitor MDM

For launch and the journey to Mercury, the MPO and the MMO will be carried as part of the Mercury Composite Spacecraft (MCS). The MCS comprises, in addition to the two orbiters, the Mercury Transfer Module (MTM), which provides solar-electric propulsion and all services not required in Mercury orbit, and the MMO Sunshield and Interface Structure

C

1

W

M

17

(MOSIF), which provides thermal protection and the mechanical and electrical interfaces for the MMO (an example of the structure is shown in Fig. 1.4, left panel). ESA is building the MTM and the MOSIF. Shortly before Mercury orbit insertion, the MTM is jettisoned from the spacecraft stack. The MPO provides the MMO with the necessary resources and services until it is delivered into its mission orbit, when control is assumed by JAXA.

The three components MTM, MPO, MMO are planned to be launched together on an Ariane 5 launch vehicle in 2015. The spacecraft will have a six year interplanetary cruise to Mercury using solar-electric propulsion and gravity assists from the Moon, Earth, Venus and eventual gravity capture at Mercury.

Arriving in Mercury orbit in January 2022, the spacecraft will have a 1-year nominal scientific life.

The main objectives of the mission are: to study the origin and evolution of a planet close to its parent star; to learn about Mercury’s form, interior, structure, geology, composition and craters; to investigate Mercury's vestigial atmosphere (exosphere), its composition and dynamics; to study Mercury's magnetised envelope (magnetosphere), structure and dynamics; to investigate the origin of Mercury's magnetic field.

The complex particle environment that surrounds Mercury is composed of thermal and directional neutral atoms (exosphere) originating via surface release and charge-exchange processes, and of ionized particles originated through photo-ionization and again by surface release processes. In order to accomplish the scientific goals, in-situ analysis of the environmental elements is necessary, and for such a purpose it was thought to build SERENA.

SERENA, Search for Exospheric Refilling and Emitted Natural Abundances (Orsini et al. 2010), is the most important instrument package for our study.

As shown in Table1.3, its instrument will include four units: two Neutral Particle Analyzers (ELENA, Emitted Low-Energy Neutral Atoms, and STROFIO, STart from a Rotating FIeld mass spectrOmeter), and two Ion Spectrometers (MIPA, Miniature Ion Precipitation Analyser and PICAM, Planetary Ion CAMera).

In particular, ELENA will investigate the neutral gas escaping from the surface of Mercury, and the related processes; STROFIO will investigate the exospheric gas composition; PICAM will measure the exo-ionosphere extension and composition, and the close-to-planet

C

1

W

M

18

magnetospheric dynamics; MIPA will study the plasma precipitation toward the surface and ions energized and transported throughout the environment of Mercury.

Finally, SERENA is an experiment able to provide information on the whole surface-exosphere-magnetosphere system, as well as on the processes involved in this system, subjected to strong interaction with the solar wind and the interplanetary medium (Milillo et al., 2005).

As will be shown in the following chapter, this work aims to provide a contribution to BepiColombo mission, in the framework of a scientific collaboration between Osservatorio Astrofisico di Catania of INAF (Istituto Nazionale di Astrofisica), and IAPS-INAF (Istituto di Astrofisica e Planetologia Spaziali) in Rome, in which operate the Principal Investigators of the experiments ISA (Italian Spring Accelerometer) and SERENA (Orsini et al., 2010) of the space mission.

In Fig. 1.4 are presented all the BepiColombo assembly and a comparison between the trajectories of its two main components and the MESSENGER satellite.

Fig. 1.4–Left panel shows BepiColombo spacecraft in its cruise configuration. On the right panel are sketched the trajectories of the MPO and the MMO that are compared to the orbit of the NASA MESSENGER spacecraft.

MESSENGER and BepiColombo scientists are making a variety of cooperative efforts involving data exchange, targeting of observations during the missions, and potentially the sharing of ground resources and collaborative scientific analysis of joint data sets.

C

1

W

M

19

As previously mentioned, space missions have given (and will give in future) an essential contribution to the knowledge of this not well known terrestrial planet.

To better understand the characteristics of terrestrial planets, in the next section the main features will be outlined.

1.3 Terrestrial planets: Mercury

Terrestrial, or telluric, or rocky planets are made up mainly of silicate rocks or metals. Silicates are a broad class of minerals containing Si. An important species of silicates are orthosilicates having the basic chemical formula SiO4-, where the oxygen atoms are

arranged at the vertices of a tetrahedron, in the center of which there is the silicon atom. Being the number of oxidation of silicon equal to +4 and that of oxygen to -2, each tetrahedron present an excess of negative charge (-4) that tends to be distributed on the four oxygen atoms: from this the presence of metal cations (Fe, Mg, K, Na, Ca), attracted by the tetrahedral anion, in the composition of these minerals. These metals have a fundamental function in binding together the various structures formed by complex silicate tetrahedrons. Aluminum is the only element that can take the place of the silicon in the center of the tetrahedron (this is called aluminosilicates).

Within the Solar System, terrestrial planets are the inner Earth-like planets closest to the Sun: Mercury, Venus, Earth, and Mars.

These planets have a solid surface and all have approximately the same type of structure: a central metallic core, mostly iron, with a surrounding silicate mantle (in molten and/or solid state) and a rocky crust. The surface is the part of the crust that is immediately in contact with the atmosphere, if this exists; it is rocky and presents a variety of topological features like canyons, craters, mountains, and volcanoes. The Moon is similar to the terrestrial planets, but it is not sure it has an iron core.

On the other hand, the much larger gas giants (or jovian planets) are composed mostly of some combination of hydrogen, helium, and water existing in various physical states. The molecules are in liquid form immediately above the rocky nucleus and in gaseous form in the outermost parts.

C

1

W

M

20

Terrestrial planets possess secondary atmospheres, generated through internal volcanism or comet impacts, unlike the gas giants, whose atmospheres are primary, captured directly from the original solar nebula.

During the formation of the Solar System, there were probably many more terrestrial planetesimals, but they likely merged or were ejected.

The uncompressed density of a terrestrial planet is the average density its materials would have at zero pressure and differs from the true average density because compression within planet cores increases their density; the average density depends on planet size as well as composition. A greater uncompressed density indicates greater metal content. A comparison between Mercury and the Earth shows that the first has an uncompressed density of 5.3 g cm−3 _{and the last of 4.4 g cm}−3_{, while the mean densities are respectively of 5.4 g cm}−3 _and

5.5 g cm−3_{. Densities generally trend towards lower values as the distance from the Sun}

increases. The main exception to this rule is the density of the Moon, which probably owes its lesser density to its unusual origin.

Fig. 1.5 – A schematic representation of the Mercury’s inner structure (a), and a mosaic photo of the Caloris Basin on Mercury’s surface taken from Mariner 10 satellite (b) (located half-way in shadow on the terminator).

Mercury has a metallic, mostly iron, core equal to 60–70% of its planetary mass and could be defined as an iron planet because of its composition, the greater density and smaller

C

1

W

M

21

radius with respect to the other terrestrial planets of comparable mass. Iron planets are believed to form in the high-temperature regions close to a star, like Mercury, and if the protoplanetary disk is rich in iron.

In Fig. 1.5 Mercury’s inner structure has been reported (a), and a picture of the surface marked by canyons, craters, mountains, and volcanoes (b).

Since Mercury surface is mainly composed of magmatic rocks a brief description about their origin and differentiation will be given in the following section.

1.4 Magmatic rocks and minerals

Magmatic rocks (also called eruptive or igneous) are formed after the solidification of magmas, i.e. masses of molten silicates containing various components (FeO, MgO, CaO, etc...) and volatile substances (water, carbon dioxide, hydrogen, methane, etc...).

These rocks form the majority in the Earth's crust, the other two categories being the sedimentary and the metamorphic ones.

Igneous rocks may form with or without crystallization, either below the surface as intrusive (plutonic) rocks or on the surface as extrusive (volcanic) rocks.

Magma can be derived from partial melts of pre-existing rocks, in either a planet's mantle or crust, and cooling. Typically, the melting is caused by one or more of three processes: an increase in temperature, a decrease in pressure, or a change in composition.

Pressure, temperature and partial pressure of the volatile magma components are important on the final process in rock formation as crystallization (slow solidification, as for granite) or amorphization (rapid solidification, as for glasses like obsidian). In Fig. 1.6 are presented a scheme of the rocks formation and two pictures of the results of the solidification process.

C

1

W

M

22

Fig. 1.6- A pictorial illustration of the rock formation process (a) and the final products due to crystallization (granite, (b)) and amorphization (obsidian, (c)).

A mineral is a naturally occurring solid chemical substance formed through biogeochemical processes, having characteristic chemical composition, highly ordered atomic structure, and specific physical properties. By comparison, a rock is an aggregate of minerals and/or mineraloids and does not have a specific chemical composition.

The nomenclature and a brief description of the most common minerals, interesting for this study, is given in Tab. 1.5.

C

1

W

M

23

Table 1.5- A selection of the silicates classification of interest for Mercury’s surface.

S

M

Characteristics Group Examples

NESOSILICATES OR ORTHOSILICATES

Isolated [SiO4]4− tetrahedrons

connected only by interstitial cations

Olivine Forsterite - Mg2SiO4 Fayalite - Fe2SiO4

INOSILICATES, OR CHAIN SILICATES

SINGLE CHAIN

Chains of silicate tetrahedrons with SiO3, 1:3 ratio

Pyroxene Enstatite - MgSiO3 Ferrosilite - FeSiO3 Diopside - CaMgSi2O6 Hedenbergite - CaFeSi2O6 Augite - (Ca,Na)(Mg,Fe,Al)(Si,Al)2O6 Jadeite - NaAlSi2O6 Wollastonite - CaSiO3 DOUBLE CHAIN

Chains of silicate tetrahedrons with Si4O11, 4:11 ratio Amphibole Tremolite Ca2Mg5Si8O22(OH)2 Actinolite Ca2(Mg,Fe)5Si8O22(OH)2 Hornblende

(Ca,Na)2-3(Mg,Fe,Al)5Si6(Al,Si)2O22(OH)

TECTOSILICATES, OR "FRAMEWORK SILICATES,"

Three-dimensional framework of silicate tetrahedrons with SiO2 or

a 1:2 ratio. They are aluminosilicates, all but the quartz group Quartz Alkali-feldspars (potassium-feldspars) Plagioclase feldspars Feldspathoids SiO2 Orthoclase - KAlSi3O8 Anorthoclase - (Na,K)AlSi3O8 Albite - NaAlSi3O8 Oligoclase (Na,Ca)(Si,Al)4O8 (Na:Ca 4:1) Anorthite - CaAl2Si2O8 Nepheline - (Na,K)AlSiO4 Sodalite - Na8(AlSiO4)6Cl2

In particular, feldspars are minerals found in intrusive and extrusive rocks and also in many types of metamorphic and sedimentary rock. They are a group of rock-forming tectosilicate minerals which make up nearly 75% of the Earth's crust. Feldspars derive from the

C

1

W

M

24

crystallization of magma containing K (Potassium-Feldspar (K-spar), KAlSi3O8), Na (albite,

NaAlSi3O8), Ca (anorthite, CaAl2Si2O8) and Ba (celsian, BaAl2Si2O8). Solid solutions

between K-feldspar and albite are called alkali feldspars. Solid solutions between albite and anorthite are called plagioclase, or plagioclase feldspar. Rocks formed almost entirely of calcic plagioclase feldspar are known as anorthosite.

A solid solution is a solid-state solution of one or more solutes in a solvent where the crystal structure of the solvent remains unchanged by addition of the solutes, and the mixture remains in a single homogeneous phase. This often happens when the two elements (generally metals) involved are close together on the periodic table.

Olivines occur in igneous rocks and as a primary mineral in certain metamorphic rocks, while pyroxenes are rock-forming silicate-chain minerals found in many igneous and metamorphic rocks.

As shown in Chapter 4, different combinations of silicates seem to make up the majority of Mercury’s surface.

Furthermore, Mercury’s location near the Sun renders it a candidate for extensive surface modifications similar to what is observed in lunar soil samples, simulated in laboratory experiments, and proposed to affect lunar soils (Vilas et al., 2008). For this reason, in the following section, the main modifications of lunar soil will be reviewed.

1.5 Mercury and the Moon

1.5.1 Lunar soil modifications

The layer of loose, heterogeneous material covering the solid rock on a celestial body is named regolith or soil. Lunar regolith is composed in part of rock and mineral fragments that were broken apart from underlying bedrock by the impact of meteorites.

When micrometeorites (< 1 mm in diameter) strike the lunar regolith, some of the regolith melts and some does not, so the final product is a glass with entrained mineral and rock fragments. These small glassy rocks composed of bits and pieces of older rocks (breccias) so formed are called agglutinates.

C

1

W

M

25

Alteration of lunar soil involve several aspects, which will be listed below (Vilas et al., 2008).

Thin nanometer-scale metallic iron (nanophase metallic Fe, npFe0_{) spheres in amorphous}

rims and lunar agglutinates are caused by meteoritic impacts, solar wind sputtering or energetic solar and cosmic rays.

In the visible and near infrared spectral region (VNIR) the result of create npFe0 _{in mafic}

(Mg, Fe) silicates is to darken the material, decrease the depth of the absorption features and to redden the overall spectrum at wavelengths >600 nm. An example of some weathered rock samples is shown in Fig. 1.7.

Fig. 1.7- (from Vilas et al., 2008)- Laboratory spectra of lunar soils (thick lines) and powdered lunar rocks (thin lines). Spectra are scaled to 1.0 at 0.35 μm. The more weathered lunar soils are redder in the visible-NIR, but bluer in the UV region, than the less weathered powdered lunar rock samples.

Weathering is observed at UV/blue wavelengths, where surface scattering dominates in non opaque minerals, (volume scattering dominates in the VNIR spectral region) and a drop in reflectance marks the transition between volume and surface scattering (150-450 nm). Furthermore, opaque materials spectra are dominated by surface scattering and are spectrally flat over the UV/blue region.

The Christiansen feature (CF) is the primary mid-infrared (MIR) spectral feature in lunar soils. It occurs near the Christiansen frequency, which is the location in frequency space where the refractive index of the material approaches the refractive index of its surrounding

C

1

W

M

26

medium. It appears as a reflectance minimum or a transmission maximum and is a good compositional indicator because its location is correlated to the location of the fundamental vibrational bands and both shifts to longer wavelengths with increasing the complexity of the silicate structure (e.g. feldspars have significantly shorter wavelength CFs than olivines). Petrologically CFs are at shorter wavelengths for felsic rocks (rich in silicate minerals, magma, and lighter elements such as silicon, oxygen, aluminum, sodium, and potassium) and shift to longer wavelengths for more mafic (Fe, Mg) rocks (as shown in Fig. 1.8); also shifts to shorter wavelengths when measured in vacuum, but are not affected by particle size, agglutinates, vitrification. The overall CF intensity is affected by temperature, but not its location (Greenhagen and Paige, 2006).

Fig. 1.8 (from Greenhagen and Paige, 2006)- Spectral emissivity of three samples of lunar soils representing the range of expected CFs. 67711 is an extremely felsic highland soil (light colored areas at higher altitudes); 61501 is a typical highland soil, and 10084 is a mafic mare soil (dark basaltic plain). The blue (A), green (B) and red (C) lines represent the location of the narrow-band filters on the lunar infrared radiometer Diviner.

In addition, lunar soil behave as terrestrial orthopyroxenes (pyroxenes which crystallize in the orthorhombic systems) with a slight deviation due to the higher Ca content in the lunar pyroxene (CF ≈8 µm) (Carli et al., 2008).

With the aim to look for any hypothetic similarity between Mercury and the Moon, in the next section will be presented a comparison of the existing data.

C

1

W

M

27

1.5.2 Similarity with the Moon?

Thermal emission measurements at microwave wavelengths (0.3-20.5 cm) indicate that Mercury’s surface is 40% more transparent than lunar highlands (light colored areas on the Moon at higher altitudes).

This observation, together with the absence of the 1µm absorption band, have led to the inference that Mercury’s crust contains lower abundances of iron plus titanium than the lunar highlands (Denevi et al., 2009).

Vernazza et al. (2010) in addition, comparing Mercury’s spectra with lunar ground based spectra and lunar soil spectra collected in laboratory, found that Mercury’s surface composition is likely to be quite different from the Moon’s.

While low-Ca iron-rich pyroxenes are the main surface components of the Moon, a Ca-rich clinopyroxene (in the hedenbergite-diopside series, CaMgSi2O6-CaFe2+Si2O6) is likely to be

a main component in Mercury’s surface. They also found that Mercury’s slope is less red than that of the Moon (in agreement with MESSENGER’s results), and composition rather than space weathering is the main cause for this difference.

An answer to the differences in the reflectance between Mercury and the Moon could be given by Lucey et al. (2011) who, using Mie theory and modifying the Hapke radiative transfer model of the effects of space weathering, found that submicroscopic iron particles (>50 nm) are much more abundant on Mercury (3.5% wt) than on Moon (0.5 % wt). Such particles darken their host material but do not redden their spectra.

These results are confirmed also in other comparisons between Mercury’s and Moon’s spectra, obtained by the MESSENGER spacecraft and spanning the wavelength range 220– 1450 nm (Holsclaw et al., 2010, Fig.1.9).

C

1

W

M

28

Fig. 1.9 (from Holsclaw et al., 2010)- Absolute reflectance of Mercury from MASCS and MDIS compared with the reflectance of the Moon at equivalent waxing phase angles acquired by the ground-based telescope Robotic Lunar Observatory (ROLO).

The brightness of an object is a function of the phase angle (the angle Sun-object-observer), which is generally smooth, except for the so-called opposition spike near 0° and when the object becomes fainter as the angle approaches 180°.

It is shared indeed that the absolute reflectance of Mercury is lower than that of the nearside waxing Moon at the same phase angle, with a spectral slope that is less steep at visible and near-infrared wavelengths, and the ratio of Mercury reflectance to that of the Moon highlights the differences between the spectral properties of the two bodies (McClintock et al., 2008).

In Fig. 1.10 the disk-averaged reflectance (observed spectral radiance of the surface divided by the radiance from a hypothetical normally illuminated Lambertian disk) of Mercury and the Moon is presented (A). In Fig. 1.10 (B) Mercury’s MUV reflectance (blue curve) exhibits a distinct departure from a linear trend (blue dashed line) at wavelengths < 300 nm, that is not observed for the Moon (red curved and dashed line), and that could be due to the presence of oxygen-transition metal charge transfer bands (OMCT, that is a transfer of a fraction of electronic charge between the molecular entities) caused by Fe 2+ _{and/or Ti} 4+_.

C

1

W

M

29

Fig. 1.10 (from McClintock et al., 2008)-A: MASCS disk-averaged reflectance of Mercury and the Moon, along with a ROLO- derived full-disk spectrum of the waxing phase (87°) of the lunar nearside, composed predominantly of highland terrain. The spectra are scaled to a value of 1.0 at 700 nm. B: Mercury’s MUV reflectance (blue curve).

The presented outcomes and the lack of an absorption feature at a wavelength near 1000 nm are interpreted as evidence for a Mercury surface composition that is low in ferrous iron within silicates but is higher in the globally averaged abundance of spectrally neutral opaque minerals than the Moon (Holsclaw et al., 2010).

Denevi et al. (2009) then concluded that, because immature materials (i.e. little exposed to space weathering) on Mercury are as much as 30% lower in reflectance than comparable materials on the Moon, Mercury’s crust seems not to be anorthositic but possibly rich in low-Fe pyroxenes, or olivine, and a spatially variable low-reflectance component.

To better study Mercury’s surface composition, in the next section some existent experiments on Mercury’s analogue materials will be treated.

C

1

W

M

30

1.6 Experiments and modeling in literature

1.6.1 Reflectance spectra and compositional heterogeneity

To support MERTIS (Mercury Thermal Infrared Imaging Spectrometer, 7-14 µm) and for cross-calibration with other instruments aboard BepiColombo and MESSENGER MASCS, Moroz et al. (2007) have done several spectral reflectance measurements (0.5-18 µm) of Mercury analogue materials together with a comparison of the thermal infrared biconical reflectance spectra (1-Rb) of the samples (slabs or particulate of various particle sizes) to

their emissivity (E) to evaluate deviations from the Kirchhoff’s law (E=1-Rb). Kirchhoff’s

law is in fact valid within certain limit for directional-hemispherical geometry, while most reflectance data on minerals and rocks are acquired at biconical geometry. A space weathering simulation experiment was also made on two FeO-poor samples of plagioclases (andesine-labradorite (Ca, Na)(Al, Si)4O8)) irradiated with nanosecond pulsed laser.

The authors found significant variations in reflectance values and absorption band contrasts with grain size in the 0.5-18 µm spectral region (Fig.1.11, left panel) and, in high-resolution NIR reflectance spectra, were recognized weak absorption features imputable to OH and/or H2O bearing weathering products (Fig. 1.11, right panel).

Fig. 1.11 (from Moroz et al., 2008)- Left panel: Biconical reflectance spectra of diopside separates. Right panel: Identification of weak actinolite (Ca2(Mg, Fe)5Si8O22(OH)2)

absorption bands (Metal-OH overtones) in NIR reflectance spectra of a diopside (CaMgSi2O6) powder.

C

1

W

M

31

Furthermore, deviations of 1-Rb from E are ≤ 3% for loose powders at ambient pressure, but

are very significant in the cases of slabs and compact fine powder (see Fig.1.12).

Fig. 1.12 (from Moroz et al., 2008)- TIR E vs. 1-Rb for andesine-labradorite unpolished slab (left panel), and loose and packed fine (<25 μm) olivine powders (right panel).

Packing effects are much less pronounced in emission than in biconical reflectance.

These effects have to be taken into account when comparing remote sensing emission spectra of planetary surface areas covered with bare rocks or compact crust with laboratory spectra at biconical geometry on slabs or simulated crust.

Finally, it is important to underline that these results were performed at ambient pressures and may not be applicable in low pressure environments.

The simulation experiment of a micrometeorite bombardment onto an andesine-labradorite sample highlight no darkening, but mild NIR (Near-InfraRed) reddening, as can be seen in Fig. 1.13 (left panel). This could indicate that the presence of FeO in the target minerals could be crucial to produce space darkening and reddening of planetary regoliths induced by micrometeorite bombardment.

Nevertheless Yamada et al. (1999) results seem to contradict these hypotheses.

Irradiating with a pulse laser beam hypersthenes ((Mg,Fe)SiO3) and enstatite (MgSiO3

-(Mg,Fe)SiO3) (with hypersthenes containing 70% more FeO than enstatite), with total

irradiated energy in unit area 240 mJmm-2_{, for 30 mJ pulse energy, was noticed that whereas}

the spectra show reddening in the infrared region, absorption band features around 1µm (ferrous iron in both olivines and pyroxenes- Strazzulla et al., 2005 and reference therein) and 2µm (pyroxenes) are not largely changed (Fig. 1.13, right panel).

C

1

W

M

32

Fig. 1.13 (Left panel: from Moroz et al., 2008)- Biconical reflectance spectra (0.5-18 μm) of non-irradiated and laser-irradiated plagioclase pellets. (Right panel: from Yamada et al., 1999)- Reflectance spectra of (top) enstatite, and (bottom) hypersthene pellet samples before and after the irradiation of pulse laser. As for enstatite and hypersthene, laser irradiation was performed repeatedly at 30mJ pulse energy. Vertical axis is reflectance (%) and horizontal axis is wavelength (nm).

So their results show that darkening and reddening are not dependent on FeO abundance. Another important step is to understand the exosphere of Mercury, because it provides information on the planet’s atomic surface composition. This study will be presented in the next section.

1.6.2 Exosphere-surface relationship

According to three MESSENGER flybys, Mercury’s exosphere is known to contain H, He, Na, K, Ca, Mg and, possibly, O (Burger et al., 2010). By extrapolating from these spacecraft atmospheric measurements, with the help of experimental results and theoretical modeling, it could be possible to deduce the major surface sources and sinks for each atmospheric component, in particular sodium, as well as to constrain the mechanisms involving thermal desorption and interactions between the surface and its space environment. These

C

1

W

M

33

interactions include bombardment of the surface by photons, solar wind ions and electrons, as well as meteorites (Dukes et al., 2011 and references therein), and will be described below

.

A different exospheric distribution of neutral Na, Ca and Mg in intensity and morphology, obtained analyzing UVVS-MASCS measurements, implies differences in the regions of the surface from which the material was derived and in the mechanism responsible for supplying each species to the exosphere.

Burger et al. (2010), developing a Monte Carlo model of the exosphere, propose Photon Stimulated Desorption (PSD) as a primary source of Na atoms from the top few nanolayers of the regolith, in the tail region (anti-sunward of the planet, between about 2 and 8 Mercury radii, formed by radiation pressure on atoms ejected from the surface). On the dayside, Na is rapidly depleted due to the high UV photon flux, while ion precipitation in regions of the surface open to the solar wind increases the diffusion rate of Na, which produces an increased flux of Na at high latitudes, detected by MASCS during the fantail observations (occurred as MESSENGER passed through Mercury’s shadow, near the closest approach to the planet) (see Fig. 1.14).

Fig. 1.14 (from Burger et al., 2010)- Comparison of Na (red), Ca (blue), and Mg (black) observations during the second MESSENGER flyby (M2). The magenta lines show model simulations of the Na exosphere. The dotted line indicates an isotropic sodium source. The dashed line shows a PSD source with the flux enhanced at high latitudes by ion precipitation.

C

1

W

M

34

However, a PSD or thermal source presume a spatial distribution that is not obtained by the exospheric Na around Mercury: as has been seen, it is at time concentrated at northern or southern high latitudes and its distribution changes significantly over short time intervals. To elucidate the contribution of sputtering by solar wind ions to the atmosphere of Mercury, Dukes et al. (2011) performed laboratory simulations of solar wind interaction with planetary surfaces irradiating with 4 keV He+ _{ions the Na-bearing tectosilicates as well as}

adsorbed Na layers deposited on albite and on olivine (a neosilicate that does not contain Na). Sodium released from the surface of Mercury can, indeed, re-deposit as a coating on different mineral surfaces by the action of gravity. The terrestrial plagioclase feldspars were chosen for their relatively high Na content and mineralogical similarity to anorthosite (lunar highland basalt): albite (NaAlSi3O8), labradorite [(Na, Ca)Si3O8], anorthoclase [(K,

Na)AlSi3O8], and olivine [(Mg, Fe)2SiO4] which contains no intrinsic Na was used as a

control.

Measurements of the sputtering of Na from these minerals were made using X-ray photoelectron spectroscopy (XPS), while secondary ion mass spectroscopy (SIMS) analyses of the sputtered flux showed that a large fraction of the Na is sputtered as ions rather than as neutral atoms. The other ions ejected from the mineral surface are Al+_{, Si}+ _{(all isotopes in}

proportion), Ca+ _{(and Ca}++_{), O}+_{, C}+_{, as well as one Na-bearing molecular ion: NaO}+ _(Fig.

1.15, left panel). The authors found that the release of Na from a mineral surface strongly depends on the abundance of Na in the soil and the irradiation history. Samples of terrestrial albite, anorthoclase and labradorite (Fig. 1.15, right panel) were studied to measure the surface depletion of intrinsic sodium (bound within the mineral structure) under ion irradiation. In all cases, the Na concentration follows a similar exponential decay, with a depletion cross section (decay constant) of 1.4 × 10-17 _cm2 _{for albite, 1.2 × 10}-17 _cm2 _in

C

1

W

M

35

Fig. 1.15 (from Dukes et al., 2011)- Left panel: SIMS spectrum induced by 4 keV He+ _on

albite that has been partially depleted of Na by high fluence irradiation. Right panel: Decrease in the surface concentration of Na relative to Si in cleaved minerals vs. fluence of 4 keV He+_.

Furthermore, few measurements of the flux and energy distribution of electrons impacting Mercury’s surface were made by Mariner 10 and MESSENGER XRS, which indicated high energy electrons impacts on the night side.

Orlando et al. (2010) performed a laboratory Electron Stimulated Desorption (ESD) experiment in an Ultra High Vacuum system (with base pressure of about 5 x 10-10 _Torr)

using pulsed (100 ns; 20-100 Hz) low energy (5-500 eV) electron beams, with fluxes of about 1010-1014 electrons/cm2s, to cryogenically cooled glasses, minerals or molecular solid targets, like Na bearing silicates. To characterize the quality and chemical composition, a FTIR (Fourier Transform Infrared Spectroscopy), temperature-programmed desorption (TPD) and optical reflection were used, while the ions produced and released were measured using a quadrupole mass spectrometer (QMS) and a ToF (time of flight) mass spectrometer. They used a 3-D hybrid model to simulate conditions of the first two flybys, to compute electron impact flux and the energy distribution. A magnetosphere of Mercury qualitatively similar to that of Earth was found, and computations for magnetic field conditions during the first flyby indicated an electron precipitation highly focused in the northern hemisphere at ~60° N latitude near noon with flux of ~1010 _{per cm}2_{s and typical}

C

1

W

M

36

predominantly at the equator, with energies of 1-5 keV and flux about the same of the first flyby, 1010 _{per cm}2_{s. From electron irradiated silicates direct yields of Na}+_{, O}+_{, H}+ _{and trace}

amount of water and Si+ _{have been determined. The relative ion abundances observed, using}

100 eV electrons, was: H+_>O+_>Na+_>Si+ _{and the ion yield showed a linear dependence on}

electron flux.

To conclude, the exospheric distribution of neutral Na, Ca and Mg could be due to photon stimulated desorption (Burger et al., 2010) or ion sputtering (Dukes et al., 2011), while electron stimulated desorption gives part of the ionized exosphere (Sprague et al., 2010; Orlando et al., 2010). Observations, experiments and modeling, regarding the exosphere, are then an indispensable tool for adding new pieces to the knowledge of the constituents of the Mercury’s surface.

However, the information available so far in this regard are not entirely complete and the literature is divided on the interpretation of some data, as for the case of sulfides, which will be introduced in Chapter 4.

Trying to summarize the literature relative to MESSENGER, ground-based and experimental data, in the next paragraph the main characteristics of Mercury’s surface and exosphere so far found will be traced.

1.7 Outline: which are, then, Mercury’s surface building blocks

Ground observations in the visible, near and medium infrared spectral region (0.4-25 µm) of Mercury have shown a heterogeneous mineralogical composition of the surface (i.e. silicates, feldspars, olivines, pyroxenes, iron oxides, which have been described in §1.4) and the presence of sodium, potassium and calcium, confirmed by the new data collected from the MESSENGER satellite.

Mercury has a generally volcanic surface with high abundances of magnesium and calcium (Nittler et al., 2011) and should consist largely of agglutinates (Vilas et al., 2008), basaltic material (Denevi et al., 2009), a small part of sulfides (0.58% mol): ZnS, CaS (oldhamite), MgS (komatiite), FeS (troilite), elemental S (Wurz, 2010; Helbert, 2007, 2012), up to 4 wt. % (Nittler et al., 2011), plagioclase feldspars (D’Amore et al., 2011; Vilas et al., 2008), (or Mg/Ca rich garnets (X3Y2SiO4), Sprague et al., 2009), thermally shocked glassy phase and